The rate of heterosexual transmission of HIV- 1 to women is increasing. This is probably taking place at mucosal sites particularly the vagina, but infection following rectal and oral sex is also possible. Protection at these sites probably requires induction of HIV 1-specific antibody secretion and cytotoxic T cells (CTL). We found that conjugates of HIV-1- derived peptides to B. abortus (V3-BA), induced HIV-1-specific serum neutralizing antibodies; and CTL which lyse HIV-1-infected target cells, in mice immunized intraperitoneally. We determined that the same constructs induce specific mucosal antibody, IgG and IgA, in the rectal and vaginal secretions of female mice following i.p. immunization. Furthermore, we demonstrated that V3-BA stimulates anti-peptide IgG and IgA responses in female mice depleted of CD4+ T cells by treatment with monoclonal anti-CD4 antibody. In order to optimize mucosal responses female BALB/c mice were immunized three times with saline, BA, V3-KLH, and V3-BA intranasally. The sera collected one week after the third immunization were tested for anti- peptide antibodies. High titer IgG (greater than 12,800) was detected following V3-KLH and V3-BA. In addition serum igA anti-peptide was generated following V3-BA immunization, suggesting that the carrier BA is capable of providing the microenvironment required for switching to IgA. The same mice were shown to have developed mucosal immunity as evidenced by the IgA anti-peptide present in their feces. Three mice from each group were sacrificed and their MLN cells were assayed for specific ELISPOTS. Mice immunized with V3-BA had 46 anti-V3 IgG secreting cells per 106 cells in MLN and spleen, respectively. These data suggest that the carrier BA, is capable of generating mucosal immunity in female mice and may do likewise in humans. Mice with targeted disruption of their MHC class II genes are deficient in T-helper cell function. This situation is similar to that occurring in HIV-1 infected persons who lose CD4+ cell function and number as the disease progresses. Unlike normal (wild-type) female mice, these mice are unable to respond to the T-cell dependent conjugate, V3-KLH. In contrast, and despite their immunodeficiency, they respond to V3-BA in a manner similar to wild-type mice. The antibody responses against the HIV-1 V3 peptide include high titer IgA. These findings show that BA as a carrier is able to generate antibody responses even in the absence of T-cell help and thus may be effective in persons that lack CD4+ T cell help as a consequence of HIV- 1 infection. The results also indicate that it is possible to elicit mucosal immune responses even in mice that lack cognate T-cell help. On the basis of the success in the murine experiments, we immunized female Rhesus macaques intramuscularly with V3-BA. Four female monkeys were immunized three times intra- muscularly at monthly intervals with saline, BA and V3-BA and tested for anti -peptide antibody responses. Two monkeys receiving V3-BA were found to generate high titer serum anti-V3 IgG antibody responses. These latter monkeys also secreted detectable anti-V3 IgG at different mucosal surfaces. Importantly, the sera from both monkeys immunized with V3-BA were able to inhibit syncytia induced by HIV-1 (MN) at levels similar to that of a human neutralizing serum provided by the NIH as a standard. Moreover, mucosal samples from were able to neutralize HIV-1 (MN) as assessed by the syncytia inhibition assay. These responses suggest that V3-BA can generate systematic and mucosal anti-HIV immune responses and suggest that vaccinations may protect humans from heterosexual transmission.